A Service of
Leibniz-Informationszentrum econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible. zbw for Economics
Rossi Júnior, José Luiz; Aguiar, Fernando
Article Understanding the evolution of the fiscal situation of the Brazilian states; 2006–2015
EconomiA
Provided in Cooperation with: The Brazilian Association of Postgraduate Programs in Economics (ANPEC), Rio de Janeiro
Suggested Citation: Rossi Júnior, José Luiz; Aguiar, Fernando (2018) : Understanding the evolution of the fiscal situation of the Brazilian states; 2006–2015, EconomiA, ISSN 1517-7580, Elsevier, Amsterdam, Vol. 19, Iss. 1, pp. 105-131, http://dx.doi.org/10.1016/j.econ.2017.06.003
This Version is available at: http://hdl.handle.net/10419/179666
Standard-Nutzungsbedingungen: Terms of use:
Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes.
Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise use the documents in public. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated licence.
https://creativecommons.org/licenses/by-nc-nd/4.0/ www.econstor.eu
HOSTED BY
Available online at www.sciencedirect.com
ScienceDirect
EconomiA 19 (2018) 105–131
Understanding the evolution of the fiscal situation of the Brazilian
states; 2006–2015
a,∗ b
José Luiz Rossi Júnior , Fernando Aguiar
a
Escola de Administra¸cão de Brasília — IDP, SGAS quadra 607 Módulo 49, Via L2 Sul, Brasília, DF, Brazil
b
Catholic University of Brasilia, Brazil
Received 24 January 2017; received in revised form 2 June 2017; accepted 5 June 2017
Available online 16 June 2017
Abstract
The text analyzes how the Brazilian states’ fiscal position evolved between 2006 and 2015, with the data revealing a clear
deterioration in state-level public finances during that period. The rating methodology developed by the Ministry of Finance is
used to show that, when comparing 2006–2008 to 2013–2015, 21 of the 26 states and the Federal District saw their fiscal position
deteriorate. The results suggest that after the global financial crisis the states failed to pursue a fiscal rule that would curb the growth
of spending in a context of falling revenue and rising debt. The study shows that, despite shrinking revenue, the states maintained the
pace of expenditure growth, particularly payroll and pension expenses. Moreover, the text shows that following the crisis, state-level
revenue would have declined by even more were it not for a substantial increase in credit inflows. While additional borrowing
enabled the states to maintain public investment in the short term, this policy showed to be unsustainable. The paper shows that
higher debt and the lack of the adjustment in public spending have a negative impact on state-level investment in the long term.
JEL classification: E61; E62; E65
Keywords: Fiscal rules; Brazil; Federalism
© 2017 The Authors. Production and hosting by Elsevier B.V. on behalf of National Association of Post-
graduate Centers in Economics, ANPEC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
After the economic stabilization in 1995, the fiscal situation of the states collapsed. In response, the federal govern-
ment adopted a program to renegotiate state-level debts, which encompassed a fiscal adjustment plan and, subsequently,
the establishment of the Fiscal Responsibility Law (FRL).
Ten years after those measures were adopted, Piancastelli and Boueri (2008) analyzed the evolution of the states’
fiscal situation and found that state-level public finances improved between 1995 and 2006. As evidence of this, the
authors noted a reduction in the ratio between the net consolidated debt and net current revenue, and improvements
in the states’ primary balances, which turned from negative (deficits) to consistently positive results (surplus). For
∗
Corresponding author.
E-mail addresses: [email protected] (J.L. Rossi Júnior), [email protected] (F. Aguiar).
https://doi.org/10.1016/j.econ.2017.06.003
1517-7580 © 2017 The Authors. Production and hosting by Elsevier B.V. on behalf of National Association of Postgraduate Centers in Economics,
ANPEC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
106 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
example, the authors show that all Brazilian states posted primary surpluses in 2004 and 2005. The same improvement
was seen in the states’ nominal balance. The authors also show that tax revenue grew in the period, particularly in
the case of the main state levy—the Goods and Services Sales Tax (ICMS). On the expenditure front, the authors
identify a drop in payroll expenses relative to GDP. On the downside, they note that the current spending — despesas
de custeio—which were not subject to any limit under the FRL—grew sharply. The authors also noted that the states
met their fiscal targets in the period by slashing investment, and most of them were unable to contain the growth of the
refinanced debt.
In contrast to the improvement that occurred in the first few years after the renegotiation process, 20 years later
the states are again in a fragile fiscal position; and a new debt refinancing program is being designed by the federal
government to alleviate their fiscal situation. The question that arises is how and why the process of improving the
state’s public accounts has been reversed. This study attempts to analyze these issues, focusing on the changes that
occurred after 2005, which weakened the states’ fiscal position.
Initially, the study analyzes the path of the states’ fiscal situation by calculating the trend of their ratings, using the
1
methodology established by the Ministry of Finance in conjunction with the FRL indicators. This is done to make
a time-series comparison of the states’ fiscal situation considering a group of indicators rather than focusing on just
one. The results show that, among the 26 states and Federal District, 21 federative entities saw their fiscal position
deteriorate when comparing 2006–2008 to 2013–2015. The figures show that borrowing is not the main cause of the
current fiscal crisis in the states. Although the federative entities’ debt ratio has risen since 2012, in most states it is
lower than in 2006. The same is true with respect to the debt service, which in 2014 recorded its lowest level relative
to state revenue.
The data show that, in the period under study, the states did not systematically abide by a fiscal rule consistent with
public debt sustainability and, hence, with the maintenance of fiscal balance. The text illustrates that, despite shrinking
revenues, the states maintained continuous growth in payroll expenses, current expenditure and pensions. These two
factors—the country’s faltering growth and the states’ failure to react to curb expenditure growth—undermined their
fiscal position, resulting in a sharp fall in the primary surplus and a further rise in the debt level.
It is interesting to note the role played by the credit obtained mainly from public sector banks and multilateral
organizations in the deterioration of the states’ fiscal position. The text shows that both the volume and number of
loans made to the states by public banks and, to a lesser extent, by multilaterals, have increased substantially since
2008. While inflows from borrowing enabled the states to cushion the impact of the fall in tax revenue, and thus avoid
having to cut public investment in the short term, it also allowed expenditure growth to continue—thereby putting off
the adjustment needed to maintain fiscal balance and restricting the states’ future investment capacity.
The rest of this paper is organized as follows. Section 2 shows the data about the trend of the primary balance
in the states, the path of states’ revenue, the dynamic of state borrowing and credit and the evolution of state-level
expenditure including payroll and pensions expenses. Section 3 analyzes the fiscal situation of the states through the
Ministry of Finance’s risk rating measure and the FRL indicators. Section 4 estimates a fiscal reaction function for
the federative entities between 2006 and 2015 analyzing whether the states followed a fiscal rule compatible with the
fiscal sustainability. Section 5 studies the determinants of state-level investment focusing on the role of credit and the
impact of other types of expenditure on states’ capital investment. Section 6 concludes with a brief discussion about
the future of fiscal policy in the federative entities.
2. Data
The data sources used are the consolidated balance sheets published by the entities, the Budget Execution Reports
(Relatório Resumido de Execu¸cão Or¸camentária — RREO), the Fiscal Management Report (Relatório de Gestão
Fiscal — RGF) in conjunction with information from the state accounting data collection system (Sistema de Coleta
de Dados Contábeis de Estados e Municípios — SISTN).
The fiscal reports present information on the consolidated financial statements, the indicators regarding the budgetary
execution, the execution of estimated revenue for the financial year, the compliance with the constitutional minimum
expenses in the areas of health and education, together with the spending limits determined by the FRL especially
1
Ministry of Finance: Orders (Portaria) MF 306/2012 and STN 543/2012.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 107
1.20% 12.00% 1.00% 10.00% 0.80% 8.00% 0.60% 6.00% 0.40% 4.00% 0.20% 2.00% 0.00% 0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 -0.20% -2.00% -0.40% -4.00%
Primary Balance (% of GDP - LeŌ) Primary Balance (% of RCL - Right)
Fig. 1. Trend of the primary balance (2006–2015).
Source: National Treasury on the basis of state data (Budget Execution Report) and Central Bank. RCL stands for net current revenue.
with respect to spending with personal, and the limits of the net consolidated debt and credit operations, among others.
The text explicitly mentions if another source of data is used. All the data covers the period between 2006 and 2015.
Always when necessary nominal variables are deflated by the consumer price index (IPCA).
2.1. The evolution of the primary balance
2
The figures shown in Table A1 indicate a decline in the states’ primary balances between 2006 and 2015. In 2006,
while three states posted primary deficits (Acre, Pará, and Sergipe), none of the three had a deficit above R$90 million.
In 2007, all states shown primary surpluses. In 2008, only one recorded a deficit (Rondônia). The year 2009 represented
a landmark in the worsening of the states’ fiscal accounts, when eight states posted primary deficits, and seven did so
in 2010. There was an improvement in 2011 when just two states achieved negative results. Nonetheless, from 2012
onwards, the states’ primary balances deteriorated significantly—the worst situation occurring in 2014, when 19 states
recorded primary deficits. In 2015, as a result of the fiscal adjustment implemented by the newly elected governors,
3
the number of deficit states fell to seven. An important fact to be considered in all state-level fiscal statistics is that
although there was a common trend of worsening fiscal accounts in the period, the deterioration was more significant
in some cases. Table A1 shows that states such as Rio de Janeiro, Rio Grande do Sul, and the Federal District suffered
sharp reductions in their primary surpluses between 2006 and 2014, and they were unable to make the adjustments
needed to obtain primary surpluses in 2015.
Fig. 1 illustrates the states’ primary balances in relation to GDP, showing a falling trend in the primary surplus since
2007 with a brief period of stabilization or improvement. The figure also shows an improvement in the surplus in 2007,
followed by an almost continuous fall until 2010. Since 2011 the overall surplus has fallen continuously to become
a primary deficit. In 2015, the states attempted to implement an adjustment which succeeded in restoring a primary surplus.4
2
For facilitating the flow of the text most of tables were placed in Appendix.
3
The above-the-line figures shown in the state accounts (Summary Budget Execution Report) are presented. A brief comparison with
the below-the-line result published by the central bank is provided in Appendix. Both concepts have problems. Many states use judicial
deposits as current revenue, which makes it difficult to interpret the concept of above the line. Another problem is the failure to record cer-
tain expenses at the right moment (such as those of previous fiscal years). For details, see http://pedrojucamaciel.com/despesas-de-exercicios-
anteriores-devo-e-nao-reconheco-pago-quando-puder/. The below-the-line concept does not capture the significant increase in remaining amounts
payable which has occurred in recent years (see http://pedrojucamaciel.com/restos-a-pagar-crescem-31-e-explicam-23-a-melhora-do-resultado-
ffiscal-dos-estados-em-2015-valor-economico-2542016/). However, irrespective of the problem and the results of payroll expenditure, in a qualitative
sense the data correctly show the fiscal situation of the states and the need for greater transparency in data disclosure.
4
Subsequent analyses used current GDP as published by the IBGE.
108 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
13.00%
12.50%
12.00%
11.50%
11.00%
10.50%
10.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Current Revenue Capital Revenue
Fig. 2. Trend of total revenue as a percentage of GDP.
Source: National Treasury on the basis of state data (Budget Execution Report), and authors’ calculations.
12.00%
10.00%
8.00%
6.00%
4.00%
2.00%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Tax Revenue ContribuƟons Property tax Agricultural Industrial Services Transfers Other
Fig. 3. Breakdown of current revenue (% of GDP).
Source: Budget Execution Report of the states, and staff calculations.
2.2. The path of states’ revenue
Fig. 2 shows the trend of the states’ total revenue relative to Brazil’s GDP since 2006. The data shows a continuous
fall from 12.7% of GDP in 2008 to 12.0% in 2011. Between 2012 and 2014 the ratio of total revenue to GDP stabilized,
despite slower economic growth. Fig. 2 clarifies what happened.
The deceleration in the country’s growth rate led to a drop in current revenue, which consists mainly of tax revenues
and government transfers. To counterbalance this, there was a substantial increase in capital revenue. Whereas in 2008,
the states’ capital revenue represented 0.34% of GDP, by 2013 it had risen to 0.98%. Thus, the robust growth of capital
revenue in 2012–2014 period made it possible to stabilize total state-level revenue at around 12.1% of GDP. In 2015,
following a reduction in capital revenue, the states’ total revenue dropped to the equivalent of 11.79% of GDP. This
makes it possible to identify credit as a significant force in 2012–2014 period preventing a steeper fall in state revenue.
Later the text provides a more detailed discussion of this source of funding.
Fig. 3 breaks down the fall in current revenue in the period into its different components. Tax revenue, which
accounts for around 60% of current revenue trended down in the period. At the height of the cyclical upswing in 2008,
the states had tax revenue equivalent to 7.67% of GDP, whereas in 2015, at its low point, tax revenues represented
6.98% of GDP. The decline in tax revenue thus accounts for about 70% of the fall in current revenue that occurred
during the period.
Fig. 3 confirms the states’ argument that the reduction in federal government transfers also played an important
role in the decline in revenue. Revenue from transfers to the states peaked at 2.90% of GDP in 2008; while the trough
in the transfer cycle occurred in 2013, at 2.48% of GDP. In 2014 and 2015 there was a slight recovery in this source
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 109
7.00%
6.00%
5.00%
4.00%
3.00%
2.00%
1.00%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
ICMS IPVA ITCD Other
Fig. 4. Trend of the components of tax revenue (% of GDP).
Source: Summary Budget Execution Report of the states, and staff calculations.
0.80%
0.70%
0.60%
0.50%
0.40%
0.30%
0.20%
0.10%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
DomesƟc Credit OperaƟons (% of GDP) External credit OperaƟons (% of GDP) Total (% of GDP)
Fig. 5. Revenue from credit operations (% of GDP).
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
of revenue, which rose to 2.59% of GDP in the latter year. Nonetheless, in most states, while important, the fall in
transfers was a secondary factor compared to the reduction in tax revenue.
As shown in Table A2, the fall in transfers mainly affected states in the North and Northeast regions, where federal
transfers account for over 60% of total revenue, compared to less than 20% in the south and southeastern states. In
2006–2015, there is no declining trend in the share of intergovernmental transfers in state revenues on average across the
states. In other words, the states do not seem to have made any effort to become less dependent on federal government
transfers.5
Fig. 4 confirms the states’ heavy reliance on the goods and services tax (ICMS), which generates about 80% of
their tax revenue. Another important point to note is that the trend of ICMS revenue is very highly correlated with
national economic activity in nearly all states, and is only less so (albeit still high) in commodity-producing states,
6
where revenue from these products is less correlated with domestic economic activity.
2.2.1. Credit, borrowing, and the fiscal dynamic of the states
Fig. 5 shows the trend of revenue from credit operations in 2006–2015 period. In 2006, in nominal terms, the states
received a total of R$3.07 billion, or 0.13% of GDP, in receipts from loan operations, with R$2.04 billion in foreign
5
A new literature studies the role of this dependency on transfers on subnationals’ fiscal discipline. See Kotia and Lledó (2016).
6
An important point to be analyzed is the impact of the structural changes that have occurred in the Brazilian economy on the ICMS tax base.
The expansion of the service sector and decline in the share of industry may be causing the states’ ICMS revenue to shrink.
110 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
35.00% 50.00% 45.00% 30.00% 40.00% 25.00% 35.00% 20.00% 30.00% 25.00% 15.00% 20.00% 10.00% 15.00% 10.00% 5.00% 5.00% 0.00% 0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
External DomesƟc Total
Fig. 6. Amount of federal government guarantees to the states (% of RCL).
Source: National Treasury (Fiscal Management Report).
loans and R$1.03 billion in domestic credit. In other words, these inflows represented only a minimal fraction of the
states’ revenues.
From 2008 on, there was an increase in total financial revenue mainly reflecting domestic loans which, from 2009
onwards outweighed foreign borrowing as the main revenue from credit operations. In 2011–2013, this process was
exacerbated by a sharp increase in loans, mainly those originating domestically, which grew from R$6.81 billion
(0.16% of GDP) in 2011, to R$24.94 billion (0.47% of GDP) in 2013. Revenue from loans then fell back to R$20.17
billion in 2015 (0.34% of GDP), but this was still above the level at the start of the period.
As a result of that dynamic, revenue from credit operations increased significantly as a share of capital revenue,
rising from 30.9% in 2006 to around 70% in 2012 and 2013, before slipping back to 64.9% in 2015—still double its
7
share at the start of the period.
Fig. 6 reveals the importance of credit to the states and its behavior over the last few years. The figure also shows
the trend of the amount of guarantees offered to the states by the federal government for domestic and external loans.
Since 2011, sovereign guaranteed loans to the states have grown vigorously as a proportion of the federal government’s
net current revenue (RCL). In 2006, 22.55% of federal government net current revenue was committed to guarantees
on loans to the states, and this level remained relatively constant until 2011, when it stood at 20.47% of the federal
government’s RCL.
Since 2011, there has been a substantial increase in federal government guarantees, which rose to 45.29% of RCL
in 2015. If the pace of growth shown in 2012–2015 were maintained for two further years, together with low revenue
growth expectations, the government would attain the limit of 60% of RCL set by the FRL. Although guarantees to
both foreign and domestic loans have increased in the period, Fig. 6 shows that, considering the sharp devaluation of
the real in 2014–2015, which raises the domestic currency value of foreign loans, the growth in guarantees occurred
mainly because of domestic operations, which rose from 12.80% of RCL in 2011 to 28.89% in 2015.
Fig. 7 shows the domestic funding sources used by the states. Up to 2011, nonbank operations such as the Export
Guarantee Funds, dominated endorsements and guarantees in the states’ borrowing operations. From 2011 onwards,
there was a substantial increase in operations with public-sector banks, which grew from 0.39% of federal government
RCL in 2010 to 16.69% in 2015. In other words, since 2011, the public-sector banks have become an important source
of funding for the states. Without the credit operations, the states’ fiscal accounts in the last few years would have been
worse than those recorded earlier.
Fig. 8 provides a more disaggregated overview of the trend of the financing of the states by public-sector banks.
Between 2002 and 2007, public-sector bank lending to the states represented less than 0.10% of GDP annually, and
never surpassed 40 annual operations, averaging 0.06% of GDP during the period and 19 operations per year. Since
2008, there was a significant increase in operations (except in 2011). Credit from public-sector banks peaked in 2013.
In that year, public-sector banks made 81 sovereign guaranteed loans to the states, for a total equivalent to 0.95% of
GDP. In 2014, there was a drastic fall in this funding source, and in the following year these operations returned to their
7
Later on, the text analyzes the impact of those operations on investment and current expenditure.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 111
18.00% 16.00% 14.00% 12.00% 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Public-Sector Banks Other Credit OperaƟons
Fig. 7. Federal government endorsement and sureties on credit operations (% of RCL).
Source: National Treasury (Fiscal Management Report).
1.00% 100
0.80% 80
0.60% 60 40 0.40% 20 0.20% 0 0.00% 20022003200420052006200720082009201020112012201320142015 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 BNDES BB CEF Other
BNDES BB CEF Other
Fig. 8. Domestic sovereign guaranteed loan operations (Volume % GDP) and (No. of operations).
Source: National Treasury
0.40% 35 0.35% 30 0.30% 25 0.25% 20 0.20% 15 0.15% 10 0.10% 5 0.05% 0 0.00% 20022003200420052006200720082009201020112012201320142015 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 BID BIRD CAF OUTROS
BID BIRD CAF OUTROS
Fig. 9. External sovereign guaranteed loan operations (volume % of GDP) and (No. of operations).
Source: National Treasury.
2008 level. All of the three public-sector banks were important in the growth of lending to the states, although in most
years the National Economic and Social Development Bank (BNDES) was these borrowers’ main funding source.
Fig. 9 shows that the multilateral banks followed the lending pattern of their public-sector counterparts, albeit at a
lower level, with an increase in the volume and number of operations in 2008 followed by a contraction after 2014.
Those changes in financing exerted a significant impact on the dynamic and composition of the states’ debt. Table A3
reports the trend of the debt of the states by region, as shown in the regionalized fiscal statistics published by the Central
Bank.
Three facts emerge from the analysis of the data presented in Table A3. Firstly, the states have proven unable to
reduce the amount of their refinanced debt. The data confirm the states’ claims that the conditions of debt refinancing
applied in 1997 did not allow them to reduce the total amount of refinancing. In all regions, the total volume of
112 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
14.00% 12.00% 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Total Expenditure Current Expenditure Capital expenditure Other expenses
Fig. 10. Trend of total expenditure (% of GDP).
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
refinanced debt did not decline in the period. The change in the debt indicator demanded by the states and recently
approved by Congress and regulated by the government is an attempt to contain this debt dynamic. Mendes (2016)
concludes that this change will not have a significant fiscal impact, and that the reduction in the present value of the
debt caused by the change in the indexing factor would realign the real interest accruing on state debts to the long-term
real interest rates prevailing in the wider economy.
Secondly, albeit on an upward trend, the renegotiated debt declined as a share of total debt in the period. The data
shows that the share of the restructured debt shrank significantly in all regions, including in the South and Southeast,
which have the highest debt levels.
Thirdly, driven by the growth in credit from public banks and multilateral institutions, there was an increase in the
share of bank debt and external debt in the states’ total borrowing. The figures show that the states’ bank debt rose
sharply during the period. In 2015, bank debt already accounted for most state borrowing in the North and Center-West
regions. Another consequence of the credit growth during the period was an increase in foreign currency borrowing. In
2015, external debt represented the largest share of state borrowing in the Northeast region, reflecting the fact that the
multilateral organizations target the poorest states in the country. Although this is positive for the states, it generates
a currency mismatch in the public accounts, which needs to be managed carefully, given the volatility of Brazil’s
exchange rate.
In short, analyzing the trend of the states’ primary surplus and revenue, it can be concluded that the post-financial-
crisis period represented a change in fiscal policy. As a response to the external negative shock that caused a fall in tax
revenue, the states increased their capital revenue through new loans obtained essentially from public banks and to a
lesser extent from multilateral lenders. This generated a new form of borrowing for the states. Although still significant
in some cases, the restructured debt lost share as a result of the increase in bank and foreign currency debt. The results
show that this increase in borrowing was not followed by an increase in the primary surplus. On the contrary, the
primary balance deteriorated, and some states even posted primary budget deficits. Thus, the states were unable to
reduce their expenditure in line with the trend of their revenue. This change in the reaction function of the subnationals
will be tested lately in the text.
2.3. The path of expenditure
Fig. 10 shows the evolution of Brazilian states’ spending as a proportion of national GDP. The absence of any clear
trend in total expenditure between 2006 and 2015 means spending did not respond systematically to the decline in
revenue during the period. Total expenditure fluctuated around 12.0% of GDP between 2006 and 2015. Starting from
12.65% in 2006, total expenditure fell in the following year, before regaining its maximum level in 2009 (12.53% of
GDP), then falling to a minimum of 11.78% in 2011, and rising again to 12.29% of GDP in 2014 and 12.23% in 2015.
The second salient feature of the expenditure trend is the significant increase in current spending since 2011, despite
total expenditure having remained relatively stable in the period. Even in a context of lower growth and, hence, falling
revenue, current expenditure rose from 10.21% of GDP in 2011 to 10.96% in 2015.
The data illustrated in Fig. 11 confirm that, between 2006 and 2009, the expenditure trend observed by Piancastelli
and Boueri (2008)—an increase in other current spending compensated by a fall in payroll and pensions expenses—was
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 113
6.00%
5.00%
4.00%
3.00%
2.00%
1.00%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Payroll and Pensions Interest and Debt Service
Other Current Expenses Other
Fig. 11. Trend of current expenditure.
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
maintained. From 2010 onwards, this trend reversed, payroll expenses and benefits started to rise sharply, while other
current expenses started to fall. But the fall in other current spending was not sufficient to neutralize the increase in
payroll and pensions expenses. While other current spending dropped by 1.12 percentage points (p.p.) from 5.74% of
GDP in 2010 to 4.62% in 2015, payroll and pensions expenses rose by 1.67 p.p. of GDP from 4.10% to 5.77%.
The data in Table A4 reveal further details of the growth in payroll expenditure between 2006 and 2015. On average,
the states increased their payroll expenditure by 11.55% per year in nominal terms, while the average inflation rate for
the period measured by the consumer price index (IPCA) was 5.90%. The Federal District led the way, with expenses
increasing by an annual average of 17.74%, followed by Pará (16.26%), Santa Catarina (16.06%) and Piauí (15.67%).
The smallest increases were recorded in the states of Paraíba (8.36%), Sergipe (8.41%) and São Paulo (8.70%). Thus,
even the states with the smallest increase in payroll expenses saw expenditure growth outpacing average inflation in 8
the period.
The results still hold when only the period 2010–2015 is considered. In this case, the average increase in expenditure
was 11.78%, compared to average inflation of 6.87%. In 2012–2014, payroll expenses rose on average by 8.47% per
year while inflation averaged 6.05%. Until 2008 the increase in payroll expenses was offset by rising revenue. When
growth stalled after 2008, the states did not adapt their payroll policy to a new growth scenario, but maintained the
pace of expenditure growth, thus exerting pressure to increase spending, causing their fiscal situation to deteriorate.
Fig. 12 helps to disentangle the growth of expenditure into active workers and pensions. The results indicate that
the two forms of expenditure contributed in a similar way to the growth of payroll expenditure in the period. Spending
on active personnel rose by 0.72 p.p. from 2.86% of GDP in 2006 to 3.58% in 2015. Spending on inactive personnel
rose from 1.58% of GDP in 2006 to 2.33% in 2015, a 0.75-point rise.
Data provided by the Brazilian Institute of Geography and Statistics (IBGE) through the Basic State Information
Survey — ESTADIC, show that the total number of state-level civil servants remained broadly stable around 3.2
million in 2012–2014, the period for which the survey is available. According to IBGE figures, with population growth
estimated at 4.6% in 2012–2014, the proportion of state-level civil servants in the population declined from 1.6% in
2012 to 1.5% in 2014.
Analysis of the data presented shows that the states find it difficult to contain the increase in expenditure on both
active and inactive personnel. Moreover, on average, the growth of payroll expenditure is not mainly attributable to
an increase in the number of civil servants employed, which averaged virtually zero, although varying between states.
Thus, wage increases above inflation and decoupled from revenue growth seem to have played a predominant role in
boosting payroll expenditure in the states.
8
There is a difference between the data presented in the Fiscal Management Report and those shown in the Fiscal Adjustment Program (PAF).
For consistency, the Fiscal Management Report data are discussed, but the results are not qualitatively different.
114 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
6.00%
5.00%
4.00%
3.00%
2.00%
1.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Total Payroll Expenses AcƟve personnel InacƟve personnel
Fig. 12. Breakdown of the increase in the expenses with personnel (% of GDP).
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
2.00%
1.50%
1.00%
0.50%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Investment Financial Investment Debt AmorƟzaƟon
Fig. 13. Capital expenses.
Source: National Treasury, using state data (Budget Execution Report), and authors’ calculations.
In the case of pensions, apart from the trend of expenditure, it is also interesting to track the deficit. Hamilton et al.
(2017) show that between 2009 and 2015 there was a continuous increase in the states’ deficit with the payment of
pensions. In 2015 the states’ deficit reached 13.2% of their net current revenue (RCL) or 1.2% of GDP up from close to
10.0% in 2009. According the authors, this increase was caused by two main factors. The number of retirees increase
by more than the number of active workers and the payment for the retirees grew accentually in the period because of
the significant increase in wages paid to active workers that took place in the period, result in line with the previous
data shown in the text. Caetano (2016) shows that the actuarial deficit of the states amounted R$ 2.4 Trillion or 43.9%
of GDP in 2014 higher than found for OECD countries.
Interesting to note some points about the spending with personal in the states. The Brazilian federalism impose
that the states are the level of government responsible for the provision of different public services like education and
public safety. Moreover, the constitution establishes a minimum expenditure on some services as education and health
and there are national laws determining minimum wages for some categories like primary school teachers. It imposes
some rigidity in the budget and makes more difficult to curb the growth of the expenditure. In addition, public workers
in these services have different retirement rules usually beneficial for these workers, leading to a more problematic
9
situation in terms of pensions.
2.3.1. Capital expenditure
Fig. 13 shows that in 2006–2010 the states spent roughly 0.40% of GDP each year on debt amortization, recording
a minimum of 0.39% of GDP in 2010, but broadly stable over the period. After 2010, this type of expenditure rose
9
For a more detailed discussion on these issues, see Caetano (2016) or Hamilton et al. (2017).
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 115
1.00%
0.80%
0.60%
0.40%
0.20%
0.00% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Debt amorƟzaƟon Interest and Debt Service Total
Fig. 14. Trend of debt service.
Source: National Treasury using state data (Summary Budget Execution Report), and authors’ calculations.
1.40% 1.20% 1.00% 0.80% 0.60% 0.40% 0.20% 0.00% -0.20% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Primary Balance (% of GDP) Investment (% of GDP)
Fig. 15. Investment and primary balance.
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
to 0.53% of GDP in 2013, before falling back to 0.42% in 2014 and 0.47% a year later. Thus, in 2011–2015, debt
amortization expenditure grew to an average of 0.47% of GDP in the period.
The relevance of debt service payments to state-level finances tends to be questioned. Fig. 14 sheds light on this
issue. Total debt service, including interest, charges, and amortization, averaged 0.92% of GDP per year in the period.
The significance of this result can be interpreted in various ways. From one perspective, the amount spent on debt
service is almost the same as total state investment, which underscores the importance of this type of expense. On the
other hand, debt service accounts for less than 10% of the states’ total expenditure. Fig. 14 shows that, until 2010, when
the states consistently posted primary surpluses, the debt interest and amortization component shrank significantly from
1.02% of GDP in 2006 to 0.82% in 2010. Thereafter, with the reduction in primary surpluses and a new cycle of state
borrowing, this downward trend disappeared.
In 2006–2015, state-level investments averaged 0.95% of GDP. This level of investment is close to the federal gov-
ernment average for the period (1.00% of GDP — Source: National Treasury), thus indicating the national importance
of public investment undertaken by the individual states.
Investment expenditure by the states is extremely volatile; and the investment rate can vary sharply from one year
to the next. For example, between 2010 and 2011, investment by the states fell from 1.27% to 0.88% of GDP, a drop
of 0.40 p.p. of GDP in a single year. A similar change occurred in 2014–2015, when investment fell from 1.07% of
GDP to 0.69%. Fig. 15 reveals one of the reasons why investment varies in this way: it is the main expense used for
fiscal adjustment in the states, so the primary balance and investment expenditure are highly correlated.
The data shown in Fig. 15 suggest that, during periods of fiscal adjustment, primary surpluses are obtained essentially
by cutting back on state-level investment. Two factors would explain this high correlation: a lack of planning by the
states, which are incapable of structuring long-term projects with stable funding sources, and the rigidity of state
budgets. As most budgetary resources finance mandatory expenses such as payroll and pensions, and a large proportion
116 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
Table 1
10
Indicators used to calculate the rating.
Variable Description Weight
I — Indebtedness (END) Ratio of consolidated public debt/net current revenue 10
11
II — Debt service/net current revenue (SDrcl) Ratio of debt service/net current revenue 9
III — Primary balance servicing the debt (RPsd) Ratio of primary balance/debt service 8
IV — Payroll and pensions expenses/net current Ratio of payroll and pensions expenses/net current revenue 7 12
revenue (DPrcl)
V — Internal savings capacity (CGPP) Ratio of current revenue minus current expenses/current revenue 4
13
VI — Share of investment in total expenditure (Pidt) Ratio of total investment/total expenditure 3
VII — Share of RPPS* contributions and Ratio of the sum of RPPS contributions and earnings/pensions expenses 2
14
earnings/pensions expenses (PCRdp)
VII — Tax revenue/current expenses (RTdc) Ratio of tax revenue/current expenses 1
Source: National Treasury.
RPPS — Regime Próprio de Previdência Social [Special Pensions Regime].
of revenue is earmarked to specific areas (the states have to spend 25% of their revenue on education, for example),
when an adjustment is necessary, investment expenses are used to make it.
Table A5 shows that, despite varying shares of net consolidated revenue (RCL) devoted to investment expenditure
(in 2015, while the State of Ceará spent 15.9% of RCL on investment expenditure, Rio Grande do Sul spent just 2.18%),
the general features of excessive volatility and sharp falls in adjustment periods apply almost everywhere. A following
section explores the longitudinal and temporal determinants of state-level investment.
3. An indicator for the states’ fiscal situation
The Ministry of Finance issued Order (Portaria) 306/2012 and Order STN 543/2012 to establish the methodology
for analyzing payment capacity, and the risk to the Union of granting endorsements and guarantees, on new domestic
or external borrowing operations by the states. The text uses the rating as an indicator of the evolution of states’ fiscal
situation.
3.1. The rating of the states
The methodology is based on the federative entities’ performance in the last three fiscal years, giving a heavier
weighting to more recent periods. A total of eight economic-financial indicators are used, each of which has a specific
weight in the calculation of each state’s credit rating. Each indicator is awarded a score between 0 and 6, where 0
indicates the left-hand tail of the indicator’s distribution and 6.0 the right-hand tail. The Order defines the two extremes
for each indicator. Intermediate values are calculated according to the indicator’s value for each state and the range in
which it varies.
The following indicators are used to calculate the rating (Table 1).
After calculating the score for each state using the relative weights for each indicator listed above, the states are
ranked by their fiscal situation and credit risk, under the following rating system (Table 2).
Table 3 shows the calculated ratings of 26 Brazilian states and the Federal District in 2006–2015. The results indicate
a clear deterioration in the federative entities’ fiscal position in that period: 20 states and the Federal District saw their
10
For a better description of some of the concepts used, see Order STN 543/2012.
11
Current revenue does not include revenue for training from the Primary Education Maintenance and Development Fund (FUNDEB).
12
Payroll and pensions expenses encompass total payroll expenditure and social contributions, minus court awards, expenses of previous fiscal
years, and employment-related indemnities and refunds.
13
The investments represent appropriations for works planning and execution. Total expenses are current expenses plus capital expenses.
14
The contributions and earnings of the RPPS are calculated as the sum of the contributions for the civil servant pension plan, contributions to the
costing of military pensions, returns on RPPS investments and financial compensation paid to the RPPS by the General Pension Regime (RGPS).
Pension expenses represent the total amount of payments in respect of retirement and other pensions, plus pensions and financial compensation to
the RGPS paid by the RPPS.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 117
Table 2
Score and fiscal situation and credit risk.
Rating Interval Fiscal situation and credit risk
≤ ≤
A+ 0.00 score 0.50 Fiscal situation excellent — Credit risk almost 0
≤
A 0.50 score ≤ 1.00 Fiscal situation very strong — Credit risk very low
A− 1.00 ≤ score ≤ 1.50
B+ 1.50 ≤ score ≤ 2.00 Fiscal situation strong — Credit risk low
B 2.00 ≤ score ≤ 2.50
B− 2.50 ≤ score ≤ 3.00 Fiscal situation good — Credit risk medium
C+ 3.00 ≤ score ≤ 3.50 Fiscal situation weak — Credit risk significant
C 3.50 ≤ score ≤ 4.00 Fiscal situation very weak — Credit risk very high
C− 4.00 ≤ score ≤ 4.50
D+ 4.50 ≤ score ≤ 5.00 Situation of fiscal imbalance
D 5.00 ≤ score ≤ 5.50
D− 5.50 ≤ score ≤ 6.00
Table 3
States’ rating.
State/period 2006–2008 2007–2009 2008–2010 2009–2011 2010–2012 2011–2013 2012–2014 2013–2015
Acre A A− B+ B B− C+ C+ C+
Alagoas C− D+ C− C− D+ C− D+ D
−
Amazonas A B+ B B+ B+ B+ B B
Amapá A A A A B+ B B− B
Bahia C+ C+ B− B B B− B B
Ceará B B+ B+ B+ B+ B+ B B−
Federal District A− A− B+ B B B B B
Espírito Santo A A A A A A− B+ B
Goiás C C C− C− C− C− C− D+
Maranhão C C C C+ C+ C C+ B−
Minas Gerais C+ C C C C− D D D+
Mato Grosso do Sul C+ C+ C C C C− C− C
Mato Grosso B− C+ B− B− B− B B B−
Pará A− A− B+ B+ B+ B+ B+ A−
Paraíba B B+ B B+ B B B− B−
Pernambuco B+ B B B− B− B− C+ B−
Piauí B B− B B− B− B− B− B−
Paraná C+ C C+ C+ C+ B C+ C+
Rio de Janeiro C C C C C C− D+ D
Rio Grande do Norte A− B+ B+ B B B− B B−
Rondônia A− B+ B+ B+ B+ B B B
Roraima A B+ B+ A− B+ B B+ B
Rio Grande do Sul D C− C− C C− D+ D D+
Santa Catarina B− B− B− B− C+ C C C
−
Sergipe B B C+ B− C+ C+ C+ C
São Paulo C+ C+ C+ C+ C+ C C− C−
Tocantins A A A− A− A− B+ B+ B+
Source: Prepared by the authors.
ratings decline, which means credit risk worsened during the period. Although the deterioration in risk rating has been
broadly constant throughout the period, it has gathered pace since 2011. Between 2006 and 2011, 16 states and the
Federal District saw their rating decline, compared to 18 states and the Federal District between 2011 and 2015.
In the three-year period 2006–2008, 16 states displayed a fiscal situation rated as excellent, very strong, or strong
(ratings A or B), with 10 classified as excellent or very strong (rating A). In 2013–2015, there are 15 states with A or
B ratings, but only one (Pará) considered excellent or very strong (rating A).
The results presented in Fig. 16, below, show the trend of the indicators used to calculate the rating between 2006
and 2015. The fiscal decline during the period is linked to the sharp fall in the primary balance that began in 2009. The
118 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
6.000
5.000
4.000
3.000
2.000
1.000
0.000 2011 2007 2007 2013 2015 2006 2008 2007 2011 2007 2013 2014 2006 2012 2014 2015 2007 2010 2012 2013 2013 2010 2011 2012 2014 2009 2010 2012 2014 2007 2010 2011 2009 2006 2007 2008 2010 2009 2010 2012 2015 2008 2010 2013 2015 2006 2014 2006 2008 2009 2014 2015 2015 2006 2008 2013 2008 2013 2015 2009 2012 2008 2010 2011 2012 2014 2009 2006 2008 2011 2009 2011 2011 2006 2012 2014 2007 2009 2013 2015
I - END II - SDrcl III - RPsd IV - DPrclV - CGPP VI - Pidt VII - PCRdp VII - RTdc
Fig. 16. The states’ rating indicators.
Source: Prepared by the authors.
figure shows that, apart from the drop in revenues owing to the weakening of economic activity, the deterioration of the
states’ fiscal situation is very closely linked to the incapacity of the states in containing the growth of expenditures in an
environment of low economic growth. In particular, the situation worsens in terms of payroll expenditure (particularly
since 2012), the capacity to generate internal savings (current expenditure), and pensions payments.
In terms of the increase in expenditures it is interesting to note the differences between changes related to cyclical
movements or economic policy and structural changes. Fig. 16 shows that, in keeping with the results reported by
Piancastelli and Boueri (2008), the states had been reducing the ratio of payroll expenditure to net current revenue
since 2006. Then, as from 2012, there is a sharp rise in the ratio, indicating that decisions of economic policy are
more likely to explain the rise in the payroll expenditure ratio. In relation to pensions, Fig. 16 shows a continuous
deterioration in pensions’ expenses; in other words, apart from issues related to the business cycle, more structural
changes would explain the increase in the spending on pensions.
Fig. 16 also reveals a fall in debt levels up to 2012, although the trend has reversed since 2013. Nonetheless, in 2015,
the debt ratio was still well below the 2006 level. Reflecting this lower debt ratio, debt service also declined relative to
net current revenue during the period.
3.2. The states, the Fiscal Responsibility Law, and the Fiscal Adjustment Program (PAF)
Apart from using the rating, the fiscal situation of the states can be analyzed through the FRL indicators. The Fiscal
Responsibility Law (Complementary Law 101, of 4 May 2000) defined national parameters to be followed relating
to the public expenditure of each Brazilian federative entity (states and municipalities). According to the National
Treasury, budget constraints would aim to preserve the federative entities’ fiscal situation, according to their annual
balance sheets, to ensure the financial health of the states and municipalities, appropriate use of resources, and a healthy
administrative legacy for future managers.
Among various obligations relating to the transparency and reporting of public expenditure, the law sets limits on
state borrowing, for total payroll expenditure, and for the volume of credit operations and guarantees, among others. The
Fiscal Adjustment Program, signed by the governors of the 25 states that refinanced their debts, specifies annual targets
for a three-year period, taking account of the trend of state finances, macroeconomic indicators for the new period,
and the fiscal policy adopted by the state governments. Fulfillment of the previous year’s targets and commitments is
evaluated each year; and the targets for a new three-year period can also be updated annually. These procedures should
be observed throughout the life of the refinancing contract.
The proposed fiscal targets presented by the states and Federal District are assessed by the Ministry of Finance,
which indicates its agreement based on technical analysis methodologies, which are the responsibility of the National
Treasury Department. These seek to preserve the solvency of the federative entity, particularly in terms of its capacity
to meet its contractual commitments. The programs necessarily contain targets for the ratio of financial debts to real
net current revenue (RCL); the primary balance, civil service expenses; the collection of internally generated revenue;
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 119
0.97 1.00 0.90 0.84 0.75 0.80 0.73 0.71 0.70 0.69 0.66 0.70 0.63 0.64 0.60 0.50 0.40 0.30 0.20 0.10 0.00
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Fig. 17. Trend of the ratio of net consolidated debt to net current revenue.
Source: National Treasury on the basis of state data (Fiscal Management Report).
48.00% 47.01%
45.22% 46.00% 44.69% 43.98% 43.07% 44.00% 42.65% 41.99% 41.24% 42.00% 40.85% 40.00% 39.24% 38.00% 36.00% 34.00%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Fig. 18. Ratio of payroll expenses to net current revenue.
Source: National Treasury on the basis of data provided by the states (Fiscal Management Report) covering the executive branch.
privatization, public utility permits or concessions; administrative and property reform, and investment expenses in
relation to RCL.
Fig. 17 shows the trend of the ratio of net consolidated debt to average net current revenue, for all states between
2006 and 2015. It confirms the occurrence of two distinct patterns in the evolution of state borrowing. The debt ratio
fell continuously between 2006 and 2011, but started to rise again in 2012. Nonetheless, the debt ratio is well below
the FRL limit of 2.00, and even after the deterioration since 2011, the mean ratio is lower than recorded in 2006, and
well below the 1995 level, as shown by Piancastelli and Boueri (2008).
Table A6 shows that nearly all states behave similarly to the mean—an improvement in the debt ratio up to 2011
followed by a deterioration or stability, with the ratio of net current debt to net current revenue falling below their
2006 levels. There are only a few exceptions to this pattern. Some states—Acre, Amapá, Ceará, Minas Gerais, Rio de
Janeiro, Roraima, Sergipe, and Tocantins—currently have a debt ratio that is worse than in 2006. States such as Mato
Grosso do Sul, Rio Grande do Norte, and Paraná lowered their borrowing ratio continuously in the period.
Table A6 also shows that, while 2011 represented the end of a period of improvement in state debt, only Rio Grande
do Sul has breached the FRL borrowing limit. States such as Rio de Janeiro and Minas Gerais display indices very
close to the legal limit. Although the debt is a problem, Fig. 18 shows that the trend of payroll expenses puts greater
pressure on the states’ fiscal situation.
In 2006, the states spent on average 41.2% of their net current revenue on payroll expenses. Following a decline
lasting until 2008, this expenditure grew again to reach a level of 47.0% by end-2015. In other words, on average,
the Brazilian states are above both the alert threshold (44.1%) and the prudential limit (46.55%), set by the LRF; and
15
they are close to the ceiling (49%) defined by the law. Table A7 shows that the trend of payroll expenses, while
15
This covers only the executive power.
120 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
reflecting the common trend of rising expenditure, varies somewhat between the states. In 2006, no state had total
payroll expenditure above the upper limit specified by the law. By 2015, however, six states were above this level
(Mato Grosso, Mato Grosso do Sul, Paraíba, Rio Grande do Norte, Rio Grande do Sul, and Tocantins). In 2006, only
two states were above the prudential limit; but the number had risen to 17 in 2015. Lastly, in the latter year only two
states (Paraná and Piauí) displayed a ratio of total payroll expenses to net current revenue below the 2006 level.
It is important to stress that those results represent a lower limit for the statistics on payroll expenditure. Accounting
and legal issues lead some states to under-report the expenses actually incurred. Although this makes it harder to obtain
a specific view for the state in question, it does not reduce the problem in aggregate. The results therefore must be seen
16
as a lower limit of the fiscal crisis affecting the states.
4. The states’ primary balance, debt sustainability and fiscal reaction function
The data presented show a deterioration in the states’ fiscal situation, particularly from 2008 onward. The figures
show that this is mainly related to expenditure growth, particularly payroll and pensions. The results suggest that the
behavior of the states following the global financial crisis was not consistent with fiscal sustainability. The states did
not respond to the reduced revenue by adapting their expenditure to the new scenario, which generated a worsening
fiscal situation and a consequent decline in the respective ratings. The following analysis studies whether there was a
change in the states’ reaction function in the period.
Woodford (2001) argues that a fiscal rule which generates a sustainable debt equilibrium requires the primary balance
to react to past changes in the debt. Bohn (2008) shows that a linear fiscal rule in which the primary balance responds
17
positively to changes in the debt is a sufficient condition to validate the government’s intertemporal constraint.
Therefore, a fiscal policy is considered sustainable once the government reacts systematically to a change in public
debt by adjusting the primary fiscal balance. In this sense, the absence of any systematic policy reaction to this shock
would cause the additionally issued debt to be uncovered by future surpluses, thus violating the no-Ponzi condition.
Therefore, in order to maintain the fiscal sustainability, the government has to react systematically to changes in the
debt by increasing the primary surplus.
18
Accordingly, a fiscal rule was estimated for the Brazilian states between 2006 and 2015, using bimonthly data
from the fiscal reports published by the states. The following equation is estimated:
Prim Balanceit = α.Prim Balanceit−1 + β.Debtit + δ.controlsit + μi + ρt + it (1)
The data includes the debt and the primary balance as a percent of the states ‘total net revenue. Some control variables
are included in the regression, as mentioned by D’Erasmo et al. (2015), Ghosh et al. (2013) and Berti et al. (2016).
These include the cyclical component of the states’ expenditure and revenue calculated using an H-P filter in order to
19
control for the effect of temporary fluctuations in the primary surplus, annual dummies to control for macroeconomic
changes (ρt) such as variations in the inflation rate and state-specific fixed effects for the panel regression (μi). As
discussed in the literature, the inclusion of the lagged dependent variable although important generates a bias if Eq. (1)
is estimated by Ordinary Least Squares (OLS) or by a fixed effects model. A first difference Arellano–Bond estimator is
used in order to estimate consistent coefficients. Besides the panel approach, a state-by-state equation is also estimated.
The results of this estimation are shown in Table 4. The results of a 2006–2015 panel study show that the states
did not behave in a way that was compatible with debt sustainability. In other words, there is no systematic relation
between the dynamic of borrowing, expenditure, and revenue. As Table 4 shows, the debt ratio, although positive,
is not statistically significant when considered over the entire study period. The results show a change in behavior
16
Different interpretations of the law by the state audit departments make it difficult to assess the real fiscal situation of the state and compliance
with the LRF. Different legal interpretations have generated decisions in 21 state audit departments, which has reduced the transparency of payroll
expenses as a proportion of total state expenditure.
17
Some studies, such as Fournier and Fall (2015), argue that a positive reaction would not be sufficient in itself; the reaction needs to be large
enough to make it possible to keep the public debt on a sustainable path. In other words, for certain levels of debt, a positive reaction may not be
large enough.
18
In the last few years, several studies have been made to estimate fiscal rules. Examples include D’Erasmo et al. (2015), Ghosh et al. (2013). See
Berti et al. (2016) for a literature review and the different methodologies used to estimate the rule.
19
Various econometric models were estimated, including nonlinear models. The results are robust to any change. All results are available.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 121
Table 4
Fiscal reaction function of the states.
Period 2006–2015 2006–2010 2011–2015
Panel
Debt 0.0049 (0.010) 0.100 (0.025)* 0.021 (0.018)
Expenditure gap −0.117 (0.017)* −0.127 (0.023)* −0.145 (0.025)*
Revenue gap 0.087 (0.034)** 0.197 (0.061)* 0.100 (0.039)**
N 1458 729 729
By state
State/variable Debt Debt Debt
Acre 0.199 (0.165) 0.229 (0.275) 0.102 (0.119)
Alagoas 0.232 (0.118)*** 0.061 (0.163) 0.514 (0.224)**
Amazonas 0.608 (0.289)** 2.34 (0.92)* −0.026 (0.498)
Amapá 0.716 (0.552) 0.727 (1.03) 0.198 (0.589)
Bahia 0.118 (0.051)** 0.129 (0.050)** −0.027 (0.084)
Ceará −0.220 (0.143) −0.174 (0.152) −0.0118 (0.316)
Federal District −0.030 (0.378) −1.31 (0.647)* 1.237 (0.587)**
Espírito Santo 0.220 (0.294) 0.394 (0.680) 0.289 (0.154)***
Goiás 0.108 (0.081) 0.102 (0.070) 0.062 (0.330)
Maranhão 0.158 (0.170) 0.157 (0.203) 0.635 (0.337)***
Minas Gerais 0.122 (0.043)* 0.124 (0.036)* 0.098 (0.067)
Mato Grosso do Sul 0.058 (0.174) 0.034 (0.218) 0.110 (0.427)
Mato Grosso 0.185 (0.142) 0.265 (0.162) 0.088 (0.737)
Pará 0.055 (0.157) 0.142 (0.139) −0.126 (0.419)
Paraíba 0.091 (0.120) 0.167 (0.144) −0.024 (0.162)
Pernambuco 0.213 (0.095)** 0.863 (0.205)* 0.095 (0.101)
Piauí −0.042 (0.153) 0.129 (0.158) −0.396 (0.193)***
Paraná 0.029 (0.052) 0.040 (0.033) −0.049 (0.313)
Rio de Janeiro 0.089 (0.091) 0.278 (0.102)** −0.151 (0.120)
Rio Grande do Norte −0.030 (0.228) −0.275 (0.481) 0.221 (0.139)
− −
Rondônia 0.029 (0.108) 0.306 (0.897) −0.082 (0.065)
Roraima 0.218 (0.196) 0.395 (0.232)*** −0.259 (0.340)
Rio Grande do Sul 0.082 (0.040)** 0.032 (0.040) 0.305 (0.118)*
Santa Catarina 0.154 (0.085)*** 0.0962 (0.097) 0.217 (0.126)***
Sergipe 0.110 (0.111) 0.255 (0.098)** −0.050 (0.091)
São Paulo 0.265 (0.093)* 0.214 (0.159) 0.078 (0.092)
Tocantins 0.111 (0.206) 0.241 (0.411) 0.569 (0.382)
N (All states) 54 27 27
Note: Estimation using the Arellano-Bond dynamic panel method. “By state” means estimation by OLS that is robust for each state. *, **, and ***
indicate statistical significance at 1%, 5%, and 10%, respectively. All variables expressed as percentages of net current revenue. Standard deviation
in parentheses. N = number of observations.
during the period. Between 2006 and 2010, the states followed a fiscal reaction function that was compatible with debt
sustainability; the borrowing coefficient is positive and statistically significant. Nonetheless, this result disappears in
2011–2015, when the coefficient becomes statistically insignificant, indicating no relation between variations in the
debt, revenue, and expenditure in this period.
The connection between these results and those presented above suggest the positive impact of adopting fiscal rules.
As noted, the period 2006–2010 saw a reduction in borrowing; and the results show that this improvement is partly
attributable to the states’ consistent behavior during the period. Against this, the rule’s lack of significance led to a
worsening of the states’ debt ratio and, hence, their fiscal position.
The results reported in Table 4 reveal the limitations of structural factors in determining the states’ primary balance.
Irrespective of the period under study, the cyclical components of state-level expenditure and revenue are important
determinants of the primary outturn. In other words, much of the state’s primary balance is determined by unexpected
movements in revenue and expenditure. This confirms the states’ dependency on exceptional revenue to construct their
122 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
primary balances, significantly complicating the federative entities’ fiscal management, given the frequency of shocks
that hit the Brazilian economy.
Although no state behaved in a manner that was consistent with sustainability, analysis of the estimation for each state
20
shows significant heterogeneity. Eight states maintained behavior consistent with the rule throughout the period. Four
of those eight (Minas Gerais, São Paulo, Rio Grande do Sul, and Alagoas) display the highest indices of indebtedness,
which might suggest that the borrowing limit established by the FRL is valid for influencing the behavior of the states
when their debt ratio is close to the limit. Thus, states whose ratio is close to the limit would tend to adopt rules to
generate consistency between the path of borrowing, expenditure, and revenue, and thus avoid reaching the limit. Most
states, where indebtedness is well within the legal limit, would not be forced into behavior that is more consistent with
debt stabilization. This fact could be analyzed for the specific design of fiscal rules that reflect each state’s situation.
Another interesting fact is that, even while obeying the fiscal rule, the most indebted states are also currently in a
delicate fiscal situation. As discussed above, while they have displayed consistent behavior, the size of the surpluses
achieved may not be sufficient to contain the increase in borrowing, which would generate a fiscal deterioration in those
states. This would confirm the hypothesis that not only the direction but also the magnitude of the states’ response is
important for maintaining fiscal balance.
The results of this section show that between 2006 and 2015 the states did not respond consistently in terms of the
debt dynamic and the primary balance, particularly in the period following the global financial crisis; and this fact, in
conjunction with the Brazilian business cycle, explains the deterioration in the states’ fiscal position. The following
sections clarify the salient facts of the period in relation to the trend of state-level public expenditure and the potential
drivers of the deterioration in state-level public accounts.
5. The role of credit and states’ investment
Previous results indicate that the increase in credit played a role in the fiscal dynamics of the states. This section
estimates the importance of credit and other government expenditures in states’ investment.
5.1. Determinants of investment
The literature identifies a variety of factors as determinants of public investment. Sturm (2001) discusses the
demographic or structural, economic and political factors likely to have an impact. The author claims that population
growth exerts pressure for an increase in investment to provide services to the larger population. Apart from that,
population growth means diseconomies of scale relative to the pre-existing capital stock, thereby fueling investment
demand. Moreover, more highly urbanized states would have greater demand for investment, because the transformation
of society would fuel an increase in demand for public services such as education and health, generating a positive
correlation between the degree of urbanization and regional investment.
In economic terms, investment may be related to the business cycle. Firstly, if governments use investment as a
countercyclical tool to absorb negative revenue shocks, there would be a negative relation between investment and
variables that proxy for the business cycle. Secondly, if the cycle means governments are more financially constrained
when making investment decisions, there would be a positive relation between investment and the cycle.
If restrictive fiscal policy measures have an impact on investment, there must be a relation between the budget deficit
and borrowing level, on the one hand, and investment by the states, on the other. Similarly, other components of public
expenditure could crowd out investment; so an increase in current expenditure, for example, could have a negative
impact on investment. Lastly, public investment can be used to influence election results, generating a positive relation
21
between investment and the electoral cycle.
Sturm (2001) also finds that public investment in a sample of 123 countries in 1970–1998 rises in periods of higher
economic growth. Thus, investment is generally not used as a countercyclical policy tool. In these countries, investment
is financed by allowing the public deficit to widen, but larger deficits in the past impede investment in the present.
20
The results for each state must be viewed cautiously, since the small size of the time series detracts from the power of the tests.
21
De Haan et al. (1996) and Sturm (1998) analyze the impact of political factors on investment.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 123
Table 5
Determinants of investment by the states 2006–2015.
Base II Fixed effects II
Investment (-1) 0.129 (1.77)*** 0.146 (2.02)**
Growth of state GDP 0.091 (2.27)** 0.076 (1.72)*** 0.079 (1.93)*** 0.084 (1.88)**
Population density −0.0025 (−0.45) 0.00012 (0.23) −0.0008 (−1.29) −0.0004 (−0.63)
Population growth 0.0239 (2.17)** 0.0138 (1.24) 0.011 (1.53) 0.015 (2.02)**
Debt −0.099 (−2.35)** −0.081 (−1.82)*** −0.027 (−2.83)* −0.023 (−2.20)**
Primary balance −0.182 (−4.69)* −0.161 (−4.69)*
Lagged primary balance 0.096 (2.75)* 0.063 (2.09)**
Current revenue 0.139 (1.96)** 0.129 (2.29)**
Current expenditure −0.180 (−3.36)* −0.095 (−2.08)**
Payroll expenses −0.074 (−1.78)*** −0.110 (−3.53)*
Domestic credit 0.225 (2.77)* 0.365 (4.71)* 0.259 (3.70)* 0.435 (6.43)*
External credit 0.131 (1.28) 0.206 (1.89)*** 0.295 (2.73)* 0.330 (2.89)*
Election 0.028 (4.69)* 0.032 (5.28)* 0.029 (4.90)* 0.033 (5.37)*
N 270 270 270 270
Note: *, **, and *** represent statistical significance at 1%, 5%, 10%, respectively. The basic regression is a dynamic panel estimated by the
Arellano–Bond method in which the fiscal variables are considered endogenous. Fixed effects represent panel estimation through the fixed effects
method. The data are described in the text.
Similarly, countries with higher levels of borrowing invest less, thereby confirming the impact of fiscal constraints on
investment.
Valila and Mehrotra (2005) analyzed the determinants of public investment for member countries of the European
Monetary Union (EMU), and found that public investment is determined by the trend of national revenue and factors
relating to the country’s fiscal sustainability. The authors claim that, in the countries of the sample, public debt growth
is associated with a continuous fall in investment, and that fiscal consolidation efforts to reduce the budget deficit and
public debt are also associated with lower investment.
This study is also germane to discussion of the actions of public-sector banks in Brazil. Firstly, the existence of such
banks is defended because their action is considered crucial for minimizing the adverse effects of economic crises.
Under this view, the cost of funding operations for the Treasury would be less than the benefits generated, considering
the multiplier effects of the investments made, and the growth of aggregate demand, employment, revenue, and tax
revenue. In addition, the public banks would play a key role in sustaining different economic segments and might
be important for regional development, in the countercyclical supply of credit and the expansion of bank services
throughout the country. Against this, the authors argue that public banks generate macroeconomic imbalances such
as higher short-term inflation and public account imbalances, while discouraging the development of other financing
22
modalities, and the capital market, and leveraging the internationalization of the country’s firms.
Development banks such as the BNDES are closely scrutinized. Bandeira-de-Mello (2015) highlights the political
involvement of development banks and the incentives given to firms. Bonomo et al. (2015) show that the growth of
public credit following the 2008 crisis benefited older firms, with lower risk, that can be characterized as free from
financial constraints. The authors find that the effect of public credit on private investment has been insignificant for
Brazilian firms.
Souza (2011) analyzed the determinants of public investment in 2005–2008, and found a positive relation between
investment and revenue obtained from borrowing. The political cycle also had an impact on public investment. Apart
from this, the author found that investment fell in the first year of a new government and increased in the final year.
The lagged primary balance also exerts a positive impact on investment, and the SELIC interest rate had a negative
effect.
Table 5 reports the estimation of the determinants of investment expenditure for the Brazilian states using a panel
spanning 2006–2015. The econometric methodology is similar than used in the previous panel estimation. The results
confirm the weak persistence of state investment. The autoregressive coefficient of investment shows that shocks to
this variable do not persist, which is reflected, as noted above, in its high variability. The results confirm the procyclical
22
Lacerda and Oliveira (2011), Bacha (2007), for Brazil; Sapienza (2004) and Dinc (2005) more generally.
124 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
nature of investment in the states, with increases (decreases) in state GDP exerting a positive (negative) effect. Table 5
also shows a (weak) relation between structural factors and state-level investment. States with higher rates of population
growth seem to invest more, which suggests that investment-demand factors can explain differential capital growth
across states.
The results in Table 5 identify the fiscal factor as the key determinant of the dynamic of investment in the states.
Those with a higher debt ratio have less investment capacity. The results confirm the importance of the primary balance
for state-level investment. A larger primary balance in the previous period frees up space for greater investment today.
The results confirm that investment is the variable used by the states to implement short-term fiscal adjustment. A better
primary balance is associated with a drop in investment. This result also holds when, instead of the primary balance, the
estimation is performed on revenue and current expenses. There is a crowding-out effect between the different forms
of expenditure, whereby an increase in current expenditure (or payroll expenses) has a negative effect on investment.
The figures also show that a relaxation of the states’ financial constraints through increased borrowing leads to an
increase in investment, thus confirming that the states are dependent on external financing to continue investing. An
increase in external and domestic credit leads to an expansion of investment by the states. Lastly, the results confirm
the use of investment as a way to influence the population’s electoral preferences; election years have a positive impact
on investment.
6. Conclusions, debt renegotiation and the future of the states’ fiscal policy
The text has shown that the states’ fiscal situation worsened in the period from 2006 to 2015. The text indicates that
the failure of public expenditure to respond to the declining revenue after 2008 generated the fiscal crisis in the states.
The fiscal policy adopted by the states after the financial crisis was unable to restrain expenditure growth, creating
significant fiscal imbalances in the states. The text shows that until recently, the fall in expenditure occurred almost
exclusively through investment cutbacks, without a major adjustment in the public accounts.
The results in the paper show that in the period leading up to the global financial crisis, the states reaped the benefits
of the stronger fiscal position represented by falling debt levels and consistent primary surpluses. This scenario provided
a more favorable setting and also fueled an increase in investment; in other words, in this period, the states succeeded
in increasing their investments mainly by using their own funding sources. In the post-global-financial-crisis period,
however, a drop in revenue, compounded by the states’ failure to react by curbing expenditure, meant a reduction in
the primary surplus and an increase in the magnitude of current expenses. Initially, this reduction in internal funding
sources was partially offset by an increase in domestic and external credit. This strategy led to an increase in debt
and did not have a persistent impact on growth. In 2015, with continuing low growth rates, a credit crunch, and the
need to reverse the fiscal framework, the states’ investment fell sharply. These results indicate that the fiscal regime
adopted after the financial crisis was unsustainable and undesirable in the long term, given its significant impact on the
country’s economic growth, since public investment by the states is just as important as federal investment, in terms
of magnitude and impact on the country’s growth potential.
The text also brings that the use of external funds in a crisis period should be implemented with caution. The paper
shows that external funding sources—public banks and multilateral agencies—played an important role, for a short
period, in maintaining public investment in the states; but the overuse of this form of financing has proven perverse
in the long term, since it postponed the necessary adjustment in public expenditure, by making it possible to stabilize
revenue. The text shows that the delay in adjustment will generate a crowding-out effect in the public accounts, which
will have an adverse impact on investment by the states. Greater current expenditure will mean less investment; and
greater indebtedness will hamper the states’ investment capacity in the future.
It is clear from the analysis that a new fiscal framework that avoids the systematic occurrence of federal government
bailouts for the states is needed. One first step is the initiative to impose a limit on expenditure growth in the states’
similar than adopted by the federal government. The imposition of such rule is welcome, given the need for the states
to contain their spending. This would help to avoid larger imbalances in the future and might lead to an improvement
in the governments’ monitoring, enforcement, and transparency system.
Various measures such as freezing civil servants’ wages and reducing monthly expenses on discretionary benefits,
would be extremely useful in the short term. Others such as the approval of fiscal responsibility laws in the states
and others that address an important set of structural changes, such as the institution of a complementary pension
regime compatible with actuarial balance rules, and the alteration of their staff statistics would also be beneficial.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 125
Also important, it is the change in the FRL, which envisages the preparation of a multiyear payroll expenditure plan,
spanning four years, together with a prohibition on wage increases in the 180 days prior to the end of the mandate of
the respective government agency or body.
This could be a good time to improve federative cooperation in terms of preparing a fiscal adjustment plan that
involves all federative entities. In relation to all entities, lasting adjustment needs to include the elimination of earmarked
revenue, which makes the budgeting process excessively rigid. Better coordination also means reviewing policies on
raising the minimum wage, and their relation to pension payments. Apart from that, a reduction in the fiscal war could
end the erosion of the tax base in the states which has been occurring over the last few years. Lastly, the country would
gain from a tax reform with the potential introduction of a value-added tax (VAT) shared between the federative entities,
which could improve the business climate for companies and, consequently, enhance business competitiveness, thereby
exerting a positive impact on growth.
Apart from the general measures mentioned, the states could take individual steps to reformulate their fiscal man-
agement as a whole. These would need to include an improvement in budgetary forecasting by putting mechanisms in
place to undertake an independent fiscal evaluation to analyze the quality of forecasts, more in-depth fiscal risk analy-
sis, enhanced transparency, particularly with regard to the disclosure and analysis of the states’ contingent liabilities,
the establishment of results-based budgeting, better alignment with international public accounting standards, better
coordination of financial execution of the major investment projects, expansion of project cost-benefit analysis, with
the implementation of a system for evaluating and monitoring results; and lastly, enhanced public expenditure auditing
and quality control mechanisms.
Appendix.
Table A1
Trend of the states’ primary balances (above the line — R$ million).
State 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
−
Acre 84.5 143.8 215.6 −238.2 −420.2 94.0 −447.9 −412.8 −329.2 −56.1
Alagoas 327.0 639.1 467.8 92.9 375.1 470.5 338.7 310.2 −278.9 713.4
Amazonas 250.2 334.3 158.4 −931.8 −322.4 96.2 291.9 −659.2 −837.1 282.4
Amapá 271.6 499.8 505.4 333.4 1174.4 734.4 −257.0 −116.8 −410.8 378.9
Bahia 1201.7 1879.9 1437.3 738.0 744.5 772.9 1278.9 300.9 1128.1 −401.1
Ceará 160.7 1134.0 1380.2 1164.6 877.2 1881.4 −202.0 637.8 −134.1 −624.1
Federal District 53.1 631.6 273.1 −415.0 35.6 11.8 −314.1 −1189.5 −514.2 −1468.3
Espírito Santo 1428.9 1433.7 2031.8 659.2 69.1 1490.0 1545.6 −316.5 −494.6 503.9
Goiás 928.8 809.8 1103.5 701.2 212.6 1974.2 1125.5 215.9 −680.6 1143.9
Maranhão 553.3 1068.4 690.1 246.2 −223.3 621.6 403.3 −308.3 −630.8 292.0
Minas Gerais 2729.2 2308.8 2971.3 1732.7 1851.1 2756.7 3068.5 −86.2 1031.0 445.7
Mato Grosso do Sul 226.6 825.6 678.1 457.4 81.1 303.6 561.1 65.8 −18.3 2004.8
Mato Grosso 526.7 544.2 785.5 120.1 721.4 769.6 698.9 −658.3 −306.7 1292.5
Pará −60.6 370.0 442.1 86.2 −195.1 989.1 1115.4 513.6 514.9 387.0
Paraíba 288.8 390.5 444.4 281.2 −205.0 452.6 −133.4 −77.7 −399.8 −32.6
Pernambuco 567.7 783.5 733.9 −331.1 220.4 −357.6 −1058.3 −1094.9 −2060.5 319.4
Piauí 219.6 473.6 308.9 −174.0 40.3 415.7 493.1 −424.9 −147.5 87.3
Paraná 928.9 917.6 1031.5 867.1 918.5 1390.4 392.2 2294.3 −934.2 1901.9
23
As discussed, the values were obtained directly from the reports supplied by the states. These have a number of discrepancies with the data
published by the STN, as shown in the technical note on the trend of the federative entities’ payroll. Nonetheless, the qualitative results are similar.
126 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
Table A1 (Continued)
State 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Rio de Janeiro 20,073.9 3491.9 4369.8 1446.5 1414.7 2600.0 −908.7 −4704.2 −7339.4 −3627.6
Rio Grande do Norte 248.9 332.6 24.9 129.0 189.3 15.0 206.5 119.7 221.2 202.2
Rondônia 242.7 269.5 −140.4 −830.7 34.1 421.2 50.9 −382.1 104.6 528.2
Roraima 327.6 341.6 145.4 −126.2 17.1 279.1 163.3 −1034.6 305.9 453.2
Rio Grande do Sul 662.8 954.2 2150.8 1627.5 1582.2 1454.1 804.6 623.9 −542.2 −1211.8
Santa Catarina 741.6 1104.0 1720.2 582.0 880.4 1259.6 869.5 295.6 −654.1 125.0
−
Sergipe −1.4 442.0 408.9 167.1 −319.2 −125.0 −73.9 95.3 229.0 −44.3
São Paulo 6803.3 5917.2 5534.1 2621.9 5152.2 6196.6 6490.0 4402.5 4589.1 7565.6
Source: National Treasury on the basis of state data (Budget Execution Report). Negative (positive) numbers stand for a primary deficit (surplus).
Table A2
Share of federal transfers in total state revenue.
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Acre 58.62% 63.00% 59.02% 47.55% 49.26% 59.40% 49.47% 53.50% 50.57% 61.52%
Alagoas 47.34% 48.32% 49.98% 44.63% 42.18% 41.91% 47.77% 42.41% 43.12% 45.46%
Amapá 70.48% 69.32% 68.80% 67.57% 67.68% 68.95% 56.30% 55.77% 59.73% 67.63%
Amazonas 28.90% 28.39% 28.97% 23.37% 24.72% 25.30% 23.48% 25.23% 25.48% 28.59%
Bahia 29.55% 32.99% 34.06% 33.62% 33.04% 34.82% 32.17% 29.56% 29.92% 29.61%
Ceará 26.29% 31.01% 32.92% 32.45% 30.19% 33.54% 34.13% 31.88% 31.19% 32.33%
Federal District 9.49% 9.70% 9.69% 9.73% 9.58% 9.68% 9.13% 8.93% 8.57% 30.28%
Espírito Santo 14.39% 14.09% 17.82% 21.26% 23.04% 24.00% 28.61% 30.48% 32.99% 31.53%
Goiás 13.09% 13.86% 14.36% 21.14% 21.39% 22.08% 21.24% 20.57% 20.38% 22.11%
Maranhão 52.48% 52.57% 54.54% 53.22% 51.21% 54.12% 51.45% 43.76% 46.96% 47.23%
Mato Grosso 23.85% 25.49% 27.45% 25.66% 24.73% 26.78% 20.30% 23.31% 25.57% 26.52%
Mato Grosso do Sul 19.15% 19.20% 19.27% 19.93% 19.77% 19.14% 19.56% 18.13% 16.96% 16.13%
Minas Gerais 10.97% 18.70% 19.69% 20.71% 20.50% 20.65% 18.96% 17.72% 19.04% 18.28%
Pará 38.00% 40.36% 41.64% 38.86% 38.12% 41.50% 36.95% 35.69% 33.98% 32.81%
Paraíba 42.02% 43.42% 46.34% 46.24% 48.27% 49.67% 47.76% 44.38% 43.99% 46.37%
Paraná 21.23% 23.21% 24.12% 24.23% 23.80% 23.85% 22.55% 19.84% 19.48% 19.33%
Pernambuco 33.47% 35.37% 36.05% 33.19% 29.91% 33.41% 29.80% 28.27% 31.11% 25.57%
Piauí 47.98% 55.55% 54.61% 46.86% 49.46% 54.12% 45.08% 49.56% 51.88% 48.91%
Rio de Janeiro 8.51% 8.60% 9.56% 10.83% 10.04% 10.05% 9.18% 7.79% 7.62% 9.07%
Rio Grande do Norte 41.19% 43.29% 45.75% 40.70% 40.92% 43.02% 40.80% 37.80% 41.22% 39.35%
Rio Grande do Sul 12.01% 19.34% 19.94% 20.85% 19.47% 20.32% 18.78% 17.53% 17.05% 17.98%
Rondônia 41.16% 42.48% 41.93% 40.88% 38.18% 39.48% 38.14% 37.99% 39.00% 43.89%
Roraima 68.29% 71.12% 63.41% 61.15% 60.27% 60.98% 46.37% 56.41% 59.05% 61.49%
Santa Catarina 21.94% 18.72% 19.97% 22.81% 24.86% 23.51% 20.27% 18.52% 18.75% 18.30%
São Paulo 9.88% 9.48% 8.91% 8.46% 8.43% 8.43% 8.19% 8.41% 9.01% 9.28%
Sergipe 44.04% 49.47% 49.74% 47.67% 44.26% 51.06% 51.15% 48.08% 48.73% 51.67%
Tocantins 51.48% 47.57% 55.24% 47.64% 47.58% 50.67% 43.78% 46.42% 43.95% 48.66%
Source: Budget Execution Report of the states, and staff calculations.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 127 0.88% 0.33% 2.19% 0.49% 0.01% 1.80% 0.00% 2.06% 0.00% 2.39% 0.26% 0.23% 3.96% 0.32% 47.66% 16.52% 32.43% 29.12% 26.84% 41.73% 36.11% 15.40% 34.41% 12.01% 11.49% 73.97% 11.89% 10.87% 69.34% 15.27% Share (%) (R$
a
2015 11,435 211 3963 7782 79 525 23,995 21,898 371 20,183 31,380 11 1352 75,195 13,419 5723 12,785 4462 1 767 37,157 65,018 0 418,522 67,294 13,534 1469 565,837 10,829 233 69,051 15,205 3946 318 99,582 thousand) Volume
1.17% 0.03% 1.67% 0.73% 0.02% 1.48% 8.53% 0.00% 1.47% 9.95% 0.00% 8.54% 2.95% 0.20% 0.22% 4.19% 0.24% 54.58% 17.77% 24.78% 33.09% 31.29% 33.40% 35.08% 17.15% 37.77% 78.36% 11.03% 72.98% 11.34% Share (%) (R$
Volume 2014 11,475 246 3737 5209 7 351 21,025 20,726 458 19,595 20,917 11 924 62,631 12,235 5982 13,171 2974 1 512 34,875 50,130 0 394,794 43,032 14,861 985 503,802 9919 201 65,621 10,197 3768 213 89,919 thousand)
1.69% 0.08% 1.71% 1.30% 0.06% 1.66% 8.65% 0.01% 2.03% 6.30% 0.05% 6.47% 3.45% 0.20% 8.05% 0.96% 9.18% 4.45% 0.36% Share (%) 57.23% 19.70% 19.59% 30.76% 36.16% 30.06% 26.71% 20.00% 42.60% 83.53% 77.01% (R$
thousand) 2013 10,622 313 3657 3636 15 317 18,560 16,524 696 19,427 16,150 34 893 53,724 8586 6429 13,695 2781 2 652 32,145 28,686 217 380,125 29,436 15,703 896 455,063 6660 797 63,746 7599 3681 298 82,781 Volume
3.73% 0.15% 1.88% 6.87% 0.14% 1.81% 8.23% 0.01% 2.01% 3.92% 0.27% 5.00% 3.98% 0.20% 7.38% 1.23% 8.26% 4.52% 0.35% 55.26% 23.37% 15.61% 27.77% 44.10% 19.31% 19.50% 24.38% 45.87% 86.63% 78.26% Share (%) (R$
2012 8543 577 3613 2414 23 290 15,460 12,531 3102 19,898 8715 62 817 45,125 5788 7239 13,616 2444 2 598 29,687 16,223 1137 358,868 20,730 16,474 845 414,277 5760 959 61,071 6446 3531 274 78,041 thousand) Volume
4.67% 0.27% 2.42% 0.25% 2.03% 4.13% 0.01% 2.13% 2.62% 0.52% 4.15% 4.48% 0.22% 5.86% 1.50% 6.12% 4.59% 0.37% 48.88% 29.58% 14.18% 26.19% 10.11% 48.62% 12.80% 12.99% 29.32% 51.43% 88.02% 81.55% Share (%) (R$
Volume 2011 5779 552 3497 1676 32 286 11,822 10,395 4012 19,301 5080 101 805 39,694 3574 8066 14,150 1136 2 585 27,513 9886 1958 332,244 15,654 16,903 825 377,470 4242 1089 59,040 4432 3324 271 72,398 thousand)
8.62% 0.46% 2.34% 0.44% 2.02% 6.26% 2.98% 0.02% 2.23% 2.33% 0.79% 3.59% 4.97% 0.23% 5.27% 1.80% 5.57% 4.61% 0.39% Share (%) 45.85% 28.95% 13.78% 21.50% 16.10% 49.21% 10.73% 35.59% 52.92% 88.09% 82.36% (R$
thousand) 2010 5428 1021 3427 1632 54 277 11,839 8331 6237 19,067 4159 169 781 38,744 1597 9082 13,506 761 5 569 25,520 8287 2807 313,191 12,752 17,677 810 355,524 3604 1234 56,316 3812 3152 264 68,382 Volume
0.85% 3.34% 0.66% 2.46% 3.51% 2.53% 0.03% 2.68% 2.19% 1.12% 2.27% 5.56% 0.29% 4.44% 2.14% 5.41% 4.53% 0.47% 35.54% 12.77% 33.96% 13.54% 15.20% 20.46% 49.53% 11.68% 40.71% 50.55% 88.57% 83.01% Share (%) (R$
thousand) 2009 3339 1200 3191 1272 80 314 9396 5462 7351 17,796 4197 238 885 35,929 844 9797 12,165 608 7 645 24,066 6942 3547 280,317 7179 17,593 931 316,509 2818 1356 52,707 3438 2879 300 63,498 Volume
1.36% 5.89% 7.34% 1.08% 4.77% 1.85% 2.89% 0.30% 4.43% 1.33% 2.39% 6.01% 0.58% 3.31% 2.31% 7.65% 4.67% 0.86% 16.81% 17.17% 41.61% 17.16% 22.82% 52.13% 11.85% 41.51% 49.02% 1.48% 88.22% 81.21% Share (%) (R$
2008 1350 1379 3341 1378 109 473 8030 2701 8395 19,177 4360 398 1753 36,784 464 10,429 12,315 726 76 1113 25,123 4718 4248 281,975 7629 19,209 1841 319,620 2167 1514 53,197 5013 3056 562 65,509 thousand) Volume
1.90% 5.63% 6.39% 9.93% 1.54% 4.54% 1.41% 2.18% 0.67% 4.10% 1.35% 1.69% 1.75% 6.65% 0.59% 3.41% 2.79% 4.65% 5.06% 0.89% 14.84% 21.58% 43.78% 12.27% 25.77% 51.82% 44.95% 46.68% 87.96% 83.19% Share (%) (R$
debt.
2007 Volume 1045 1519 3082 864 134 396 7040 2254 9086 18,270 3502 544 1599 35,255 332 10,568 10,975 513 158 23,510 3847 4803 249,780 4980 18,873 283,966 1575 46,961 2626 2859 505 56,449 thousand) 2015.
Bank state-level
Union Union Union Union Union 8.727/1993) 9.496/1997) 8.727/1993) 9.496/1997) 8.727/1993) 9.496/1997) 8.727/1993) 9.496/1997) 8.727/1993) 9.496/1997)
September
of
the the the the the
of
(Law (Law (Law (Law (Law (Law (Law (Law (Law (Law
debt debt debt 964 debt 1683 debt
region with with with with with as
Central
region region
debt debt debt debt debt
A3
region region debts debts debts debts debts debt debt debt debt debt 1923
Data a Renegotiation External Other Renegotiation External Other Renegotiation Renegotiation External Other Renegotiation External Other Renegotiation External Other Renegotiation Restructured Total Renegotiation Restructured Total Restructured Total Renegotiation Restructured Total Renegotiation Restructured Total Northeast Center-west Southeast South Bank Bank Bank Bank Bank North Table Source: Composition
128 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 Average annual increase 9.27% 8.99% 10.07% 9.81% 9.54% 9.39% 17.74% 10.95% 10.62% 12.34% 12.28% 14.51% 11.88% 16.26% 8.36% 12.24% 11.22% 15.67% 9.24% 13.62% 9.64% 9.56% 12.89% 16.06% 8.70% 8.41% 12.71%
2196.77 4013.20 1757.44 6004.24 9091.78 6758.12 7185.02 7800.20 6279.54 4835.48 4387.29 6279.32 3135.20 1580.11 4204.91 3976.30 2015 18,213.54 19,672.02 11,388.75 39,278.12 10,601.15 21,737.08 12,160.37 19,155.80 23,985.17 12,484.90 80,815.59
2145.66 3739.48 2134.87 5658.02 8305.44 6688.67 6132.72 6693.84 5154.56 9289.60 4596.94 3994.65 5448.66 2917.82 1341.07 4101.00 3129.37 2014 15,884.20 10,960.63 10,323.61 33,097.98 21,391.19 11,263.66 18,215.79 21,611.47 11,334.23 73,613.02
1903.52 3305.97 1843.59 5038.88 7739.57 9582.58 5889.58 9270.67 4934.17 5831.68 4662.78 8314.94 4112.69 2588.23 5269.53 2777.31 1316.62 9750.61 3765.09 3038.29 2013 14,294.85 27,189.68 14,278.58 10,215.24 18,620.32 18,920.11 68,196.36
1852.83 3054.80 1547.81 5011.66 7493.88 8665.66 5198.76 8530.47 4832.70 5225.33 4241.64 7518.50 3959.22 9240.05 2240.17 3752.80 2619.54 1173.20 8861.00 3670.76 2691.38 2012 12,842.56 24,382.34 14,465.29 14,411.02 16,615.81 40,618.85
776.48 1481.69 2312.24 1181.95 4354.40 7155.51 7779.18 4655.27 7389.88 3353.13 4614.96 2695.21 6145.95 3247.33 8153.46 2084.00 3235.86 8426.29 2261.05 7831.81 2344.89 2415.40 2011 11,377.36 21,874.99 12,924.24 12,507.92 37,040.93
774.67 1360.02 2404.92 1092.47 3717.56 6523.91 5707.73 3050.53 6746.61 3326.09 3982.61 2424.41 4000.45 3234.96 7253.61 1825.67 2810.29 7616.66 1881.65 4324.71 3417.52 1972.28 2010 10,173.89 18,920.48 10,856.81 11,397.13 34,021.88
23 917.71 738.83 1271.40 2426.60 3438.63 9144.03 5660.66 4964.26 2491.21 5580.10 3389.05 3491.92 2101.63 3520.62 2699.41 9536.37 6338.71 1580.32 9751.02 2548.85 6725.82 1661.95 3723.73 2858.49 1627.05 2009 16,058.96 31,019.13 Million).
(R$
913.37 680.91 1130.53 2275.25 3075.45 8059.37 4944.99 5406.69 2274.52 5096.83 2929.43 3109.30 1884.65 3165.75 2233.83 8709.96 5562.52 1389.15 9375.28 2276.84 1439.36 3433.40 2412.27 1358.53 2008 16,674.56 11,001.32 29,253.04
expenditure
994.24 763.66 613.62 1980.22 2589.25 7031.96 4145.81 4276.13 2780.81 4786.06 2503.04 2616.74 1703.03 2747.39 2162.02 7815.08 4638.68 1300.81 8487.97 2072.65 1310.51 3058.50 2069.12 1375.13 2007 14,268.35 10,157.24 33,447.23 pensions
and
payroll
905.38 673.36 470.09 1697.06 2355.64 7321.07 3704.30 3843.47 2389.92 4150.98 2243.99 2449.57 1620.04 2350.34 2166.62 6850.58 4198.95 1023.46 7913.86 1751.36 9557.43 1257.91 2815.98 1874.78 1201.95 2006 12,780.45 35,082.33
nominal
Sul
Norte Sul
do
states’
do do
the
Santo
District
Janeiro
Gerais
of
Catarina A4 Grosso Grosso
Paulo de Grande Grande
Alagoas Amapá Amazonas Bahia Ceará Federal Espírito Goiás Maranhão Mato Mato Minas Pará Paraíba Paraná Pernambuco Piauí Rio Rio Rio Rondônia Roraima Santa São Sergipe Tocantins Table Acre Trend
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 129
Table A5
Ratio of state-level investment to net consolidated revenue.
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Acre 33.17% 18.46% 25.88% 41.38% 41.40% 19.34% 24.71% 24.05% 25.87% 9.68%
Alagoas 8.20% 5.24% 10.46% 15.53% 20.14% 10.01% 14.77% 15.32% 21.25% 8.26%
Amapá 10.84% 7.82% 12.21% 11.06% 8.54% 10.35% 8.09% 14.22% 10.15% 11.34%
Amazonas 19.08% 16.21% 19.79% 27.64% 22.90% 18.00% 15.44% 23.30% 20.06% 7.68%
Bahia 9.21% 6.56% 8.55% 9.06% 11.78% 9.11% 8.49% 8.83% 9.82% 8.42%
Ceará 27.36% 9.83% 13.67% 23.62% 33.67% 24.50% 17.07% 16.74% 24.14% 15.89%
Federal District 10.17% 8.93% 10.92% 13.48% 9.07% 7.58% 10.91% 13.79% 9.39% 3.65%
Espírito Santo 13.09% 11.90% 11.78% 15.74% 19.34% 12.34% 10.67% 13.40% 10.19% 5.19%
Goiás 5.94% 5.77% 8.48% 8.91% 11.20% 4.18% 4.88% 10.17% 15.15% 8.59%
Maranhão 12.98% 14.50% 18.75% 17.25% 17.04% 11.94% 12.11% 14.47% 17.29% 9.30%
Mato Grosso 15.93% 14.45% 17.35% 23.98% 15.40% 12.94% 25.80% 32.93% 26.93% 12.84%
Mato Grosso do Sul 6.80% 3.53% 12.13% 8.15% 14.38% 13.81% 9.64% 13.25% 15.76% 6.16%
Minas Gerais 12.28% 11.85% 12.74% 12.31% 12.04% 8.85% 7.87% 9.90% 8.64% 6.17%
Pará 17.02% 7.84% 13.77% 10.35% 14.62% 5.30% 7.26% 8.96% 9.87% 8.10%
Paraíba 8.00% 6.37% 8.82% 10.14% 10.78% 7.49% 10.41% 13.58% 15.64% 10.45%
Paraná 12.20% 6.35% 6.76% 6.53% 7.96% 3.85% 5.96% 7.05% 5.63% 3.22%
Pernambuco 8.61% 5.76% 7.41% 10.55% 14.64% 14.24% 14.69% 16.63% 13.77% 5.43%
Piauí 10.86% 7.10% 11.29% 19.06% 16.88% 10.40% 12.74% 19.97% 14.81% 7.89%
Rio de Janeiro 6.53% 5.30% 5.51% 9.44% 14.96% 11.95% 13.08% 14.72% 16.65% 13.00%
Rio Grande do Norte 11.24% 6.11% 6.50% 10.22% 8.89% 4.57% 6.14% 5.02% 8.83% 7.39%
Rio Grande do Sul 4.61% 2.64% 3.57% 3.44% 9.19% 4.02% 4.14% 5.06% 5.07% 2.18%
Rondônia 9.88% 9.94% 16.72% 24.11% 16.21% 10.82% 9.41% 15.94% 10.23% 9.40%
Roraima 11.83% 10.58% 15.55% 14.10% 28.99% 11.69% 15.52% 16.62% 18.33% 6.08%
Santa Catarina 8.59% 7.01% 7.53% 10.39% 9.18% 7.00% 6.69% 7.80% 11.98% 11.11%
São Paulo 5.54% 5.35% 8.51% 11.88% 12.65% 8.83% 7.02% 10.28% 9.32% 6.26%
Sergipe 9.51% 4.36% 6.37% 6.54% 10.83% 7.61% 7.07% 4.46% 8.36% 5.40%
Tocantins 25.76% 24.77% 29.18% 22.68% 22.48% 11.80% 10.97% 12.69% 15.49% 5.42%
Average 12.42% 9.06% 12.23% 14.72% 16.12% 10.46% 11.17% 13.67% 14.02% 7.94%
Source: National Treasury using state data (Budget Execution Report), and authors’ calculations.
Table A6
Ratio of net current debt to net current revenue for all states, 2006–2015.
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Acre 0.496 0.411 0.284 0.365 0.537 0.504 0.584 0.685 0.739 0.969
Alagoas 2.216 2.002 1.971 1.809 1.617 1.479 1.504 1.464 1.539 1.697
Amazonas 0.333 0.190 0.128 0.237 0.271 0.194 0.153 0.220 0.308 0.046
Amapá 0.108 0.096 0.040 0.110 0.180 0.120 0.178 0.262 0.412 0.274
Bahia 1.023 0.824 0.719 0.629 0.521 0.464 0.491 0.471 0.399 0.594
Ceará 0.604 0.383 0.235 0.172 0.277 0.294 0.277 0.295 0.426 0.628
Federal District 0.327 0.191 0.160 0.173 0.181 0.159 0.100 0.161 0.207 0.252
Espírito Santo 0.337 0.190 0.103 0.083 0.172 0.137 0.150 0.207 0.270 0.314
Goiás 1.867 1.611 1.403 1.285 1.299 1.086 1.021 0.924 0.898 0.986
Maranhão 1.151 0.914 0.750 0.687 0.645 0.474 0.411 0.382 0.463 0.600
Minas Gerais 1.891 1.878 1.763 1.795 1.823 1.818 1.745 1.834 1.790 1.987
Mato Grosso do Sul 1.811 1.483 1.153 1.142 1.216 1.131 1.054 1.024 0.982 0.917
Mato Grosso 1.098 0.941 0.700 0.541 0.553 0.420 0.305 0.346 0.424 0.450
Pará 0.436 0.345 0.283 0.239 0.286 0.194 0.108 0.103 0.100 0.120
Paraíba 0.729 0.600 0.482 0.342 0.357 0.249 0.263 0.267 0.370 0.414
Pernambuco 0.666 0.531 0.425 0.430 0.386 0.383 0.458 0.528 0.579 0.622
Piauí 0.847 0.779 0.602 0.602 0.540 0.568 0.505 0.586 0.609 0.570
Paraná 1.308 1.163 1.187 1.122 0.909 0.758 0.598 0.521 0.581 0.485
Rio de Janeiro 1.725 1.735 1.604 1.629 1.562 1.461 1.651 1.538 1.782 1.978
Rio Grande do Norte 0.297 0.221 0.192 0.175 0.207 0.172 0.148 0.148 1.202 0.090
130 J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131
Table A6 (Continued)
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Rondônia 0.765 0.644 0.504 0.558 0.541 0.496 0.486 0.712 0.621 0.607
Roraima 0.103 0.125 0.128 0.311 0.041 0.051 0.196 0.375 0.184 0.118
Rio Grande do Sul 2.536 2.538 2.345 2.195 2.140 2.138 2.181 2.086 2.093 2.272
Santa Catarina 1.089 0.903 0.774 0.608 0.629 0.457 0.406 0.479 0.451 0.530
Sergipe 0.571 0.423 0.218 0.267 0.333 0.477 0.528 0.551 0.571 0.689
São Paulo 1.895 1.707 1.630 1.511 1.529 1.457 1.539 1.419 1.478 1.678
Tocantins 0.128 0.085 0.099 0.110 0.164 0.206 0.206 0.256 0.327 0.400
Source: National Treasury on the basis of data provided by the states.
Table A7
Ratio of total payroll expenditure to net current revenue (Fiscal Management Report).
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Acre 0.481 0.462 0.459 0.461 0.460 0.443 0.463 0.450 0.445 0.488
Alagoas 0.459 0.469 0.458 0.510 0.452 0.464 0.479 0.477 0.497 0.485
Amazonas 0.411 0.380 0.387 0.462 0.419 0.422 0.414 0.436 0.456 0.480
Amapá 0.400 0.410 0.392 0.418 0.426 0.429 0.451 0.445 0.450 0.434
Bahia 0.423 0.427 0.431 0.468 0.432 0.444 0.446 0.453 0.455 0.476
Ceará 0.390 0.399 0.382 0.408 0.409 0.400 0.420 0.435 0.428 0.459
Federal District 0.412 0.368 0.424 0.434 0.434 0.461 0.450 0.450 0.469 0.468
Espírito Santo 0.320 0.331 0.297 0.356 0.383 0.365 0.363 0.424 0.433 0.434
Goiás 0.439 0.483 0.431 0.461 0.473 0.428 0.449 0.448 0.463 0.484
Maranhão 0.355 0.365 0.343 0.386 0.408 0.349 0.369 0.392 0.387 0.438
Minas Gerais 0.446 0.464 0.458 0.462 0.486 0.389 0.407 0.415 0.435 0.479
Mato Grosso do Sul 0.438 0.409 0.353 0.397 0.434 0.410 0.404 0.401 0.386 0.552
Mato Grosso 0.393 0.332 0.364 0.398 0.412 0.436 0.527 0.438 0.464 0.502
Pará 0.441 0.459 0.431 0.456 0.445 0.444 0.448 0.479 0.459 0.473
Paraíba 0.439 0.439 0.453 0.516 0.573 0.467 0.498 0.480 0.493 0.509
Pernambuco 0.422 0.416 0.417 0.448 0.436 0.426 0.452 0.449 0.462 0.462
Piauí 0.428 0.430 0.401 0.428 0.433 0.442 0.446 0.477 0.442 0.427
Paraná 0.449 0.440 0.423 0.451 0.459 0.467 0.467 0.472 0.468 0.433
Rio de Janeiro 0.275 0.262 0.239 0.270 0.268 0.263 0.296 0.296 0.333 0.335
Rio Grande do Norte 0.469 0.503 0.465 0.501 0.488 0.483 0.489 0.489 0.385 0.525
Rondônia 0.398 0.384 0.335 0.385 0.398 0.393 0.438 0.466 0.436 0.444
Roraima 0.287 0.307 0.331 0.357 0.370 0.349 0.415 0.469 0.394 0.473
Rio Grande do Sul 0.413 0.423 0.381 0.405 0.386 0.404 0.426 0.434 0.457 0.492
Santa Catarina 0.440 0.399 0.370 0.371 0.418 0.418 0.465 0.467 0.479 0.483
Sergipe 0.429 0.429 0.402 0.456 0.453 0.468 0.483 0.487 0.580 0.478
São Paulo 0.429 0.415 0.408 0.413 0.394 0.403 0.424 0.420 0.440 0.463
Tocantins 0.447 0.426 0.360 0.441 0.477 0.471 0.488 0.517 0.509 0.517
Source: National Treasury on the basis of data provided by the states (Fiscal Management Report).
References
Bacha, E. (2007) Bancos públicos: o que fazer? In: Pinheiro, A.; Oliveira Filho, L. de (Org.). Mercado de capitais e bancos públicos: análises e
experiências comparadas. Rio de Janeiro: Contra Capa e ANBID, p. 265–269.
Bandeira-de-Mello, R., 2015. Betting on the Winner: The Effect of Local Elections on Corporate Political Activity Outcomes. Mimeo, FGV São
Paulo School of Business.
Berti K., Colesnic E., Desponts C., Pamies S., Sail E. (2016). “Fiscal Reaction Functions for European Countries”. European Comission Discussion
Paper 028. April 2016.
Bohn, H., 2008. The sustainability of fiscal policy in the United States. In: Neck, Reinhard, Sturm, Jan-Egbert (Eds.), Sustainability of Public Debt.
MIT Press, Cambridge, Massachusetts.
Bonomo, M., Brito, R., Martins, B., 2015. Macroeconomic and financial consequences of the after crisis government-driven credit expansion in
Brazil. J. Int. Money Finance 55, 11–34.
Caetano, M. (2016). Solvencia fiscal de longo prazo dos regimes próprios de previdência dos estados e municípios. Texto para Discussão 2195 — IPEA.
J.L. Rossi Júnior, F. Aguiar / EconomiA 19 (2018) 105–131 131
D’Erasmo, P., Mendoza, E., Zhang, J. (2015). “What is a sustainable debt?” NBER Working Paper 21574.
De Haan, J., Sturm, J.E., Sikken, B., 1996. Government capital formation: explaining the decline. Weltwirtschaftliches Archiv 132 (1), 55–74.
Dinc, S., 2005. Politicians and banks; political influences on government-owned banks in emerging markets. J. Financ. Econ. 77, 453–479.
Fournier, J., Fall, F. (2015). Government debt and fiscal framework, Annex 2 the government debt limit. OECD Working paper,
ECO/CPE/WP1(2015)7/ANN2.
Ghosh, A.R., Kim, J., Mendoza, E., Ostry, J., Qureshi, M., 2013. Fiscal fatigue, fiscal space and debt sustainability in advanced economies. Econ.
J. 123, F4–F30.
Hamilton, C., Almeida, V., Caldas, L., Sá, J., Martins, F., Brito, A., 2017. A dinâmica do déficit dos regimes próprios de previdência dos estados
brasileiros nos anos 2006–2015. Carta de Conjuntura 34, IPEA.
Kotia, A., Lledó, V. (2016). “Do subnational fiscal rules foster fiscal discipline? New empirical evidence from Europe”. IMF Working Paper 16/84.
Lacerda, A., Oliveira, A. (2011). “O papel dos bancos públicos no Brasil: uma análise da atuac¸ão do BNDES nos 2000.” IV Encontro Internacional
da Associa¸cão Keynesiana Brasileira (AKB).
Mendes, M. (2016). “O impacto fiscal das propostas de refinanciamento da dívida dos estados.” Boletim Legislativo Nr. 44 Federal Senate,
March/2016.
Piancastelli, M., Boueri, R. (2008). “Dívida dos Estados 10 anos Depois.” Texto para discussão Nr. 1366, IPEA.
Sapienza, P., 2004. The effects of government ownership on Bank lending. J. Financ. Econ. 72 (2), 357–384.
Souza, A., 2011. Determinantes do investimento público no Brasil; Uma Abordagem em painel dinâmico para os estados brasileiros. Mimeo, Federal
University of Ceará.
Sturm, J., 1998. Public Capital Expenditure in OECD Countries. The Causes and Consequences of the Decline in Public Capital Spending. Edward
Elgar.
Sturm, J., 2001. Determinants of Public Capital Spending in Less-developed Countries. Mimeo, University of Groningen & CESifo Munich.
Valila, T., Mehrotra, A. (2005). Evolution and determiants of public Investment in Europe. European Investment Bank Economic and Financial
Report 2005/01.
Woodford, M. (2001). Fiscal requirements for price stability. NBER Working Paper 8072.